Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add kellykampen/agent-skills --skill finchgit clone --depth 1 https://github.com/kellykampen/agent-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/kellykampen/agent-skills/finch)<a href="https://agentmods.dev/skills/kellykampen/agent-skills/finch"><img src="https://agentmods.dev/badge/skills/kellykampen/agent-skills/finch/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kellykampen/agent-skills/finch"><img src="https://agentmods.dev/badge/skills/kellykampen/agent-skills/finch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00183 | $0.02254 |
| Opus 5 | $0.00092 | $0.01127 |
| Sonnet 5 | $0.00037 | $0.00451 |
| Haiku 4.5 | $0.00018 | $0.00225 |
Grade A, and why
finch scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
finch (X/Twitter CLI + MCP server)
finch is a Twitter/X CLI built for both humans and agents, backed by the official X API v2
with bring-your-own-keys (BYOK) OAuth 2.0 PKCE auth. It ships a bundled MCP server (finch mcp,
stdio transport) exposing the exact same functionality as native MCP tools, not shelled-out CLI
calls — so an agent harness gets deterministic JSON in/out instead of parsing terminal output.
Auth — a human must set this up first, not you
Finch uses OAuth 2.0 Authorization Code + PKCE. The human running you needs an X Developer Portal app configured as a Native App / public client (no client secret), with this redirect URI registered:
http://127.0.0.1:8765/callback
They then run the interactive browser flow:
finch auth
finch auth opens the system browser, X shows a consent page, and Finch captures the
authorization code on a short-lived local callback server at http://127.0.0.1:8765/callback.
The Client ID can be passed via --client-id, the FINCH_OAUTH2_CLIENT_ID environment
variable, or an interactive masked prompt. After one live validation call succeeds, credentials
are written to ~/.finch/config at 0600 permissions.
As an agent, you cannot do this step. finch auth is an interactive, human-only,
browser-based OAuth 2.0 flow. If a Finch tool call fails with an auth error (exit code 3 on
the CLI, {code: "AUTH_ERROR", ...} from an MCP tool), tell the human operator to run
finch auth themselves on their own machine — don't attempt to work around it, guess at
credentials, or ask the user to paste keys into chat.
Command / tool surface
Every CLI command has a matching MCP tool, both calling the same underlying logic (no re-implementation between the two surfaces):
| Capability | CLI command | MCP tool |
|---|---|---|
| Post a top-level tweet | finch post "<text>" [--media <path> ...] [--alt <text> ...] |
post_tweet |
| Reply to a tweet | finch reply <id-or-url> "<text>" |
reply_tweet |
| Post a thread (chained replies) | finch thread "<text1>" "<text2>" ... or finch thread --file <path> |
post_thread |
| Delete a post | finch delete <id-or-url> |
delete_tweet |
| List bookmarks | finch bookmark list [-n] [--folder <id>] |
list_bookmarks |
| List bookmark folders | finch bookmark folders |
list_bookmark_folders |
| Add / remove a bookmark | finch bookmark add <id-or-url> [--folder <id>] / finch bookmark rm <id-or-url> |
add_bookmark / remove_bookmark |
| Create a bookmark folder | finch bookmark folder new "<name>" |
create_bookmark_folder |
| Draft an article | finch article draft "<title>" <markdown-path> [--cover <path>] |
article_draft |
| Publish an article draft | finch article publish <draft_id> |
article_publish |
| Create and publish an article | finch article post <markdown-path> --title "<title>" [--cover <path>] |
article_post |
| Home reverse-chronological timeline | finch timeline [-n] |
get_timeline |
| Recent search (~7 days on free/basic tiers) | finch search "<query>" [-n] |
search_tweets |
| A user's recent posts | finch user-posts <username> [-n] |
get_user_posts |
| A user's profile | finch user <username> |
get_user_profile |
| Fetch one post by id/URL | finch show <id-or-url> |
get_tweet |
| Like / unlike a post | finch like / finch unlike <id-or-url> |
like_tweet / unlike_tweet |
| Repost / undo repost | finch repost / finch unrepost <id-or-url> |
repost_tweet / unrepost_tweet |
| Follow / unfollow a user | finch follow / finch unfollow <username> |
follow_user / unfollow_user |
| Own identity | finch whoami |
whoami |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 142 lines · 183 tokens per session scan A 21b215399ede
finch is a skill published in the GitHub repository kellykampen/agent-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 183 tokens to every session and 2,254 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…